Insects and other pests cost farmers billions of dollars annually in crop losses and in the expense of keeping these pests under control. The losses caused by insect pests in agricultural production environments include decreases in crop yield, reduced crop quality, and increased harvesting costs. Insect pests are also a burden to vegetable and fruit growers, to producers of ornamental flowers, and to home gardeners and homeowners.
Cultivation methods, such as crop rotation and the application of high levels of nitrogen fertilizers, have partially addressed problems caused by agricultural pests. However, various demands on the utilization of farmland restrict the use of crop rotation. In addition, overwintering traits of some insects are disrupting crop rotations in some areas.
Thus, synthetic chemical insecticides are relied upon most heavily to achieve a sufficient level of control. However, the use of synthetic chemical insecticides has several drawbacks. For example, the use of these chemicals can adversely affect many beneficial insects. Target insects have also developed resistance to some chemical pesticides. Furthermore, rain and improper calibration of insecticide application equipment can result in poor control. The use of insecticides often raises environmental concerns such as contamination of soil and water supplies when not used properly, and residues can also remain on treated fruits and vegetables. Working with some insecticides can also pose hazards to the persons applying them. Stringent new restrictions on the use of pesticides and the elimination of some effective pesticides could limit effective options for controlling damaging and costly pests.
The replacement of synthetic chemical pesticides, or combination of these agents with biological pesticides, could reduce the levels of toxic chemicals in the environment. Some biological pesticidal agents that are now being used with some success are derived from the soil microbe Bacillus thuringiensis (B.t.). While most B.t. strains do not exhibit pesticidal activity, some B.t. strains produce proteins that are highly toxic to pests, such as insects, and are specific in their toxic activity. Genes that encode δ-endotoxin proteins have been isolated. Other species of Bacillus also produce pesticidal proteins.
Höfte and Whiteley classified B.t. crystal proteins into four major classes (Höfte, H., H. R. Whiteley [1989] Microbiological Reviews 52(2):242-255). The classes were CryI (Lepidoptera-specific), CryII (Lepidoptera- and Diptera-specific), CryIII (Coleoptera-specific), and CryIV (Diptera-specific). The discovery of strains specifically toxic to other pests has been reported. For example, CryV and CryVI have been proposed to designate a class of toxin genes that are nematode-specific.
The 1989 nomenclature and classification scheme of Höfte and Whiteley for crystal proteins was based on both the deduced amino acid sequence and the activity spectrum of the toxin. That system was adapted to cover 14 different types of toxin genes divided into five major classes. The 1989 nomenclature scheme became unworkable as more and more genes were discovered that encoded proteins with varying spectrums of pesticidal activity. Thus, a revised nomenclature scheme was adopted, which is based solely on amino acid identity (Crickmore et al., 1998, Microbiology and Molecular Biology Reviews 62:807-813).
Recombinant DNA-based B.t. products have been produced and approved for use. In addition, with the use of genetic engineering techniques, various approaches for delivering these toxins to agricultural environments are being perfected. These include the use of plants genetically engineered with toxin genes for insect resistance and the use of stabilized intact microbial cells as toxin delivery vehicles. Thus, isolated Bacillus toxin genes are becoming commercially valuable.
B.t. protein toxins were initially formulated as sprayable insect control agents. A relatively more recent application of B.t. technology has been to isolate and transform plants with genes that encode these toxins. Transgenic plants subsequently produce the toxins, thereby providing insect control. See U.S. Pat. Nos. 5,380,831; 5,567,600; and 5,567,862 to Mycogen Corporation. Transgenic B.t. plants are quite efficacious, and usage is predicted to be high in some crops and areas.
There are some obstacles to the successful agricultural use of Bacillus (and other biological) pesticidal proteins. Certain insects can be refractory to the effects of Bacillus toxins. Insects such as boll weevils, black cutworm, and Helicoverpa zea, as well as adult insects of most species, heretofore have demonstrated no significant sensitivity to many B.t. δ-endotoxins.
Another potential obstacle is the development of resistance to B.t. toxins by insects. The potential for wide-spread use of B.t. plants has caused some concern that resistance management issues may arise more quickly than with traditional sprayable applications. While a number of insects have been selected for resistance to B.t. toxins in the laboratory, only the diamondback moth (Plutella xylostella) has demonstrated resistance in a field setting (Ferre, J. and Van Rie, J., Annu. Rev. Entomol. 47:501-533, 2002).
Resistance management strategies in B.t. transgene plant technology have become of great interest. Several strategies have been suggested for preserving the ability to effectively use B. thuringiensis toxins. These strategies include high dose with refuge, and alternation with, or co-deployment of, different toxins (McGaughey et al. (1998), “B.t. Resistance Management,” Nature Biotechnol. 16:144-146), as in a natural bacterium, for example.
Thus, there remains a great need for developing additional genes that can be expressed in plants in order to effectively control various insects. In addition to continually trying to discover new B.t. toxins (which is becoming increasingly difficult due to the numerous B.t. toxins that have alredy been discovered), it would be quite desirable to discover other bacterial sources (distinct from B.t.) that produce toxins that could be used in transgenic plant strategies.
The relatively more recent efforts to clone insecticidal toxin genes from the Photorhabdus/Xenorhabdus group of bacteria present potential alternatives to toxins derived from B. thuringiensis. The genus Xenorhabdus is taxonomically defined as a member of the Family Enterobacteriaceae, although it has certain traits atypical of this family. For example, strains of this genus are typically nitrate reduction negative and catalase negative. Xenorhabdus has only recently been subdivided to create a second genus, Photorhabdus, which is comprised of the single species Photorhabdus luminescens (previously Xenorhabdus luminescens) (Boemare et al., 1993 Int. J. Syst. Bacteriol. 43, 249-255). This differentiation is based on several distinguishing characteristics easily identifiable by the skilled artisan. These differences include the following: DNA-DNA characterization studies; phenotypic presence (Photorhabdus) or absence (Xenorhabdus) of catalase activity; presence (Photorhabdus) or absence (Xenorhabdus) of bioluminescence; the Family of the nematode host in that Xenorhabdus is found in Steinernematidae and Photorhabdus is found in Heterorhabditidae); as well as comparative, cellular fatty-acid analyses (Janse et al. 1990, Lett. Appl. Microbiol. 10, 131-135; Suzuki et al. 1990, J. Gen. Appl. Microbiol., 36, 393-401). In addition, recent molecular studies focused on sequence (Rainey et al. 1995, Int. J. Syst. Bacteriol., 45, 379-381) and restriction analysis (Brunel et al., 1997, App. Environ. Micro., 63, 574-580) of 16S rRNA genes also support the separation of these two genera.
The expected traits for Xenorhabdus are the following: Gram stain negative rods, white to yellow/brown colony pigmentation, presence of inclusion bodies, absence of catalase, inability to reduce nitrate, absence of bioluminescence, ability to uptake dye from medium, positive gelatin hydrolysis, growth on Enterobacteriaceae selective media, growth temperature below 37° C., survival under anaerobic conditions, and motility.
Currently, the bacterial genus Xenorhabdus is comprised of four recognized species, Xenorhabdus nematophilus, Xenorhabdus poinarii, Xenorhabdus bovienii and Xenorhabdus beddingii (Brunel et al., 1997, App. Environ. Micro., 63, 574-580). A variety of related strains have been described in the literature (e.g., Akhurst and Boemare 1988 J. Gen. Microbiol., 134, 1835-1845; Boemare et al. 1993 Int. J. Syst. Bacteriol. 43, pp. 249-255; Putz et al. 1990, Appl. Environ. Microbiol., 56,181-186, Brunel et al., 1997, App. Environ. Micro., 63,574-580, Rainey et al. 1995, Int. J. Syst. Bacteriol., 45, 379-381).
Photorhabdus and Xenorhabdus spp. are Gram-negative bacteria that entomopathogenically and symbiotically associate with soil nematodes. These bacteria are found in the gut of entomopathogenic nematodes that invade and kill insects. When the nematode invades an insect host, the bacteria are released into the insect haemocoel (the open circulatory system), and both the bacteria and the nematode undergo multiple rounds of replication; the insect host typically dies. These bacteria can be cultured away from their nematode hosts. For a more detailed discussion of these bacteria, see Forst and Nealson, 60 Microbiol. Rev. 1 (1996), pp. 21-43. Unfortunately, as reported in a number of articles, the bacteria only had pesticidal activity when injected into insect larvae and did not exhibit biological activity when delivered orally.
Xenorhabdus and Photorhabus bacteria secrete a wide variety of substances into the culture medium. See R. H. ffrench-Constant et al. 66 AEM No. 8, pp. 3310-3329 (August 2000), for a review of various factors involved in Photorhabdus virulence of insects.
It has been difficult to effectively exploit the insecticidal properties of the nematode or its bacterial symbiont. Thus, proteinaceous agents from Photorhabdus/Xenorhabdus bacteria that have oral activity are desirable so that the products produced therefrom could be formulated as a sprayable insecticide, or the genes encoding said proteinaceous agents could be isolated and used in the production of transgenic plants.
There has been substantial progress in the cloning of genes encoding insecticidal toxins from both Photorhabdus luminescens and Xenorhabdus nematophilus. Toxin-complex encoding genes from P. luminescens were examined first. See WO 98/08932. Parallel genes were more recently cloned from X. nematophilus. Morgan et al., Applied and Environmental Microbiology 2001, 67:20062-69. WO 95/00647 relates to the use of Xenorhabdus protein toxin to control insects, but it does not recognize orally active toxins. WO 98/08388 relates to orally administered pesticidal agents from Xenorhabdus. U.S. Pat. No. 6,048,838 relates to protein toxins/toxin complexes, having oral activity, obtainable from Xenorhabdus species and strains.
Four different toxin complexes (TCs)—Tca, Tcb, Tcc and Tcd—have been identified in Photorhabdus spp. Each of these toxin complexes resolves as either a single or dimeric species on a native agarose gel but resolution on a denaturing gel reveals that each complex consists of a range of species between 25-280 kDa. The ORFs that encode the typical TCs from Photorhabdus, together with protease cleavage sites (vertical arrows), are illustrated in FIG. 5. See also R. H. ffrench-Constant and Bowen, 57 Cell. Mol. Life Sci. 828-833 (2000).
Genomic libraries of P. luminescens were screened with DNA probes and with monoclonal and/or polyclonal antibodies raised against the toxins. Four tc loci were cloned: tca, tcb, tcc and tcd. The tca locus is a putative operon of three open reading frames (ORFs), tcaA, tcaB, and tcaC, transcribed from the same DNA strand, with a smaller terminal ORF (tcaZ) transcribed in the opposite direction. The tcc locus also is comprised of three ORFs putatively transcribed in the same direction (tccA, tccB, and tccC). The tcb locus is a single large ORF (tcbA), and the tcd locus is composed of two ORFs (tcdA and tcdB); tcbA and tcdA, each about 7.5 kb, encode large insect toxins. TcdB has some level of homology to TcaC. It was determined that many of these gene products were cleaved by proteases. For example, both TcbA and TcdA are cleaved into three fragments termed i, ii and iii (e.g. TcbAi, TcbAii and TcbAiii). Products of the tca and tcc ORFs are also cleaved. See FIG. 5. See also R. H. ffrench-Constant and D. J. Bowen, Current Opinions in Microbiology, 1999, 12:284-288.
Bioassays of the Tca toxin complexes revealed them to be highly toxic to first instar tomato hornworms (Manduca sexta) when given orally (LD50 of 875 ng per square centimeter of artificial diet). R. H. ffrench-Constant and Bowen 1999. Feeding was inhibited at Tca doses as low as 40 ng/cm2. Given the high predicted molecular weight of Tca, on a molar basis, P. luminescens toxins are highly active and relatively few molecules appear to be necessary to exert a toxic effect. R. H. ffrench-Constant and Bowen, Current Opinions in Micriobiology, 1999, 12:284-288.
None of the four loci showed overall similarity to any sequences of known function in GenBank. Regions of sequence similarity raised some suggestion that these proteins (TcaC and TccA) may overcome insect immunity by attacking insect hemocytes. R. H. ffrench-Constant and Bowen, Current Opinions in Microbiology, 1999, 12:284-288.
TcaB, TcbA and TcdA all show amino acid conservation (˜50% identity), compared with each other, immediately around their predicted protease cleavage sites. This conservation between three different Tc proteins suggests that they may all be processed by the same or similar proteases. TcbA and TcdA also share ˜50% identity overall, as well as a similar predicted pattern of both carboxy- and amino-terminal cleavage. It was postulated that these proteins might thus be homologs of one another. Furthermore, the similar, large size of TcbA and TcdA, and also the fact that both toxins appear to act on the gut of the insect, may suggest similar modes of action. R. H. ffrench-Constant and Bowen, Current Opinions in Microbiology, 1999, 12:284-288.
Deletion/knock-out studies suggest that products of the tca and tcd loci account for the majority of oral toxicity to lepidopterans. Deletion of either of the tca or tcd genes greatly reduced oral activity against Manduca sexta. That is, products of the tca and tcd loci are oral lepidopteran toxins on their own; their combined effect contributed most of the secreted oral activity. R. H. ffrench-Constant and D. J. Bowen, 57 Cell. Mol. Life. Sci. 831 (2000). Interestingly, deletion of either of the tcb or tcc loci alone also reduces mortality, suggesting that there may be complex interactions among the different gene products. Thus, products of the tca locus may enhance the toxicity of tcd products. Alternatively, tcd products may modulate the toxicity of tca products and possibly other complexes. Noting that the above relates to oral activity against a single insect species, tcb or tcc loci may produce toxins that are more active against other groups of insects (or active via injection directly into the insect haemocoel—the normal route of delivery when secreted by the bacteria in vivo). R. H. ffrench-Constant and Bowen, Current Opinions in Microbiology, 1999, 12:284-288.
The insect midgut epithelium contains both columnar (structural) and goblet (secretory) cells. Ingestion of tca products by M. sexta leads to apical swelling and blebbing of large cytoplasmic vesicles by the columnar cells, leading to the eventual extrusion of cell nuclei in vesicles into the gut lumen. Goblet cells are also apparently affected in the same fashion. Products of tca act on the insect midgut following either oral delivery or injection. R. H. ffrench-Constant and D. J. Bowen, Current Opinions in Microbiology, 1999, 12:284-288. Purified tca products have shown oral toxicity against Manduca sexta (LD50 of 875 ng/cm2). R. H. ffrench-Constant and D. J. Bowen, 57 Cell. Mol. Life Sci. 828-833 (2000).
WO 99/42589 and U.S. Pat. No. 6,281,413 disclose TC-like ORFs from Photorhabdus luminescens. WO 00/30453 and WO 00/42855 disclose TC-like proteins from Xenorhabdus. WO 99/03328 and WO 99/54472 (and U.S. Pat. Nos. 6,174,860 and 6,277,823) relate to other toxins from Xenorhabdus and Photorhabdus. 
While the exact molecular interactions of the TCs with each other, and their mechanism(s) of action, are not currently understood, it is known, for example, that the Tca toxin complex of Photorhabdus is toxic to Manduca sexta. In addition, some TC proteins are known to have “stand alone” insecticidal activity, while other TC proteins are known to potentiate or enhance the activity of the stand-alone toxins. It is known that the TcdA protein is active, alone, against Manduca sexta, but that TcdB and TccC, together, can be used (in conjunction with TcdA) to greatly enhance the activity of TcdA. TcbA is the other main, stand-alone toxin from Photorhabdus. The activity of this toxin (TcbA) can also be greatly enhanced by TcdB- together with TccC-like proteins.
PhotorhabdusPhotorhabdusstrain W14SomehomologyTC proteinnomenclatureto:TcaAToxin CTccATcaBTccBTcaCTcdBTcbToxin BTccAToxin DTcdA N terminusTccBTcdA C terminusTccCTcdAToxin ATccA + TccBTcdBTcaC
Some Photorhabdus TC proteins have some level of sequence homology with other Photorhabdus TC proteins. As indicated above, TccA has some level of homology with the N terminus of TcdA, and TccB has some level of homology with the C terminus of TcdA. Furthermore, TcdA is about 280 kDa, and TccA together with TccB are of about the same size, if combined, as that of TcdA. Though TccA and TccB are much less active on SCR than TcdA, TccA and TccB from Photorhabdus strain W14 are called “Toxin D.” “Toxin A” (TcdA), “Toxin B” (Tcb or TcbA), and “Toxin C” (TcaA and TcaB) are also indicated above.
Furthermore, TcaA has some level of homology with TccA and likewise with the N terminus of TcdA. Still further, TcaB has some level of homology with TccB and likewise with the N terminus of TcdA. TcdB has a significant level of similarity to TcaC.
Relatively recent cloning efforts in Xenorhabdus nematophilus also appear to have identified novel insecticidal toxin genes with homology to the P. luminescens tc loci. See, e.g., WO 98/08388 and Morgan et al., Applied and Environmental Microbiology 2001, 67:20062-69. In R. H. ffrench-Constant and D. J. Bowen Current Opinions in Micriobiology, 1999, 12:284-288, cosmid clones were screened directly for oral toxicity to another lepidopteran, Pieris brassicae. One orally toxic cosmid clone was sequenced. Analysis of the sequence in that cosmid suggested that there are five different ORF's with similarity to Photorhabdus tc genes; orf2 and orf5 both have some level of sequence relatedness to both tcbA and tcdA, whereas orf1 is similar to tccB, orf3 is similar to tccC and orf4 is similar to tcaC. Importantly, a number of these predicted ORFs also share the putative cleavage site documented in P. luminescens, suggesting that active toxins may also be protealytically processed.
There are five typical Xenorhabdus TC proteins: XptA1, XptA2, XptB1, XptC1, and XptD1. XptA1 is a “stand-alone” toxin. XptA2 is the other TC protein from Xenorhabdus that has stand-alone toxin activity. XptB1 and XptC1 are the Xenorhabdus potentiators that can enhance the activity of either (or both) of the XptA toxins. XptD1 has some level of homology with TccB.
XptC1 was known to have some level of similarity to TcaC. The XptA2 protein of Xenorhabdus was known to have some degree of similarity to the TcdA protein. XptB1 has some level of similarity to TccC.
The finding of somewhat similar, toxin-encoding loci in these two different bacteria is interesting in terms of the possible origins of these virulence genes. The X. nematophilus cosmid also appears to contain transposase-like sequences whose presence may suggest that these loci can be transferred horizontally between different strains or species of bacteria. A range of such transfer events may also explain the apparently different genomic organization of the tc operons in the two different bacteria. Further, only a subset of X. nematophilus and P. luminescens strains appear toxic to M. sexta, suggesting either that different strains lack the tc genes or that they carry a different tc gene compliment. Detailed analysis of both a strain and toxin phylogeny within, and between, these bacterial species should help clarify the likely origin of the toxin genes and how they are maintained in different bacterial populations. R. H. ffrench-Constant and Bowen, Current Opinions in Microbiology, 1999, 12:284-288.
TC proteins and genes have more recently been described from other insect-associated bacteria such as Serratia entomophila, an insect pathogen. Waterfield et al., TRENDS in Microbiology, Vol. 9, No. 4, April 2001.
In summary, toxin complex proteins from P. luminescens and X. nematophilus appear to have little homology to previously identified bacterial toxins and should provide useful alternatives to toxins derived from B. thuringiensis. Although they have similar toxic effects on the insect midgut to other orally active toxins, their precise mode of action remains obscure. Future work could clarify their mechanism of action.
Bacteria of the genus Paenibacillus are distinguishable from other bacteria by distinctive rRNA and phenotypic characteristics (C. Ash et al. (1993), “Molecular identification of rRNA group 3 bacilli (Ash, Farrow, Wallbanks and Collins) using a PCR probe test: Proposal for the creation of a new genus Paenibacillus,” Antonie Van Leeuwenhoek 64:253-260). Some species in this genus are known to be pathogenic to honeybees (Paenibacillus larvae) and to scarab beetle grubs (P. popilliae and P. lentimorbus) P. larvae, P. popilliae, and P. lentimorbus are considered obligate insect pathogens involved with milky disease of scarab beetles (D. P. Stahly et al. (1992), “The genus Bacillus: insect pathogens,” p. 1697-1745, In A. Balows et al., ed., The Procaryotes, 2nd Ed., Vol. 2, Springer-Verlag, New York, N.Y.).
A crystal protein, Cry18, has been identified in strains of P. popilliae and P. lentimorbus. Cry18 has scarab and grub toxicity, and has about 40% identity to Cry2 proteins (Zhang et al., 1997; Harrison et al., 2000).
TC proteins and lepidopteran-toxic Cry proteins have very recently been discovered in Paenibacillus. See U.S. Ser. No. 60/392,633 (Bintrim et al.), filed Jun. 28, 2002.
Although some Xenorhabdus TC proteins were found to “correspond” (have a similar function and some level of sequence homology) to some of the Photorhabdus TC proteins, the “corresponding” proteins share only about 40% (approximately) sequence identity with each other. This is also true for the more recently discovered TC proteins from Paenibacillus (those proteins and that discovery are the subject of co-pending U.S. Ser. No. 60/392,633).
In light of concerns about insects developing resistance to a given pesticidal toxin, and in light of other concerns—some of which are discussed above, there is a continuing need for the discovery of new insecticidal toxins and other proteins that can be used to control insects.